Abstract
Ancient documents are often subject to background damage. Examples of background damage containing various quality are uneven background, ink bleed, or ink bleed and expansion of spots. Image processing provides a variety of approaches for dealing with this deterioration in the quality of manuscripts to make them readable. Unlike DIBCO dataset, Jawi-Malay manuscripts are among the ancient manuscripts that have undergone varying degrees of complex damage and distortions. State of the art methods unable to improve their image and readability. To deal with these challenges, this paper proposes a new adaptive binarization method which consists of several steps beginning with global enhancement and local adaptive thresholding, and ending with postprocessing. For the purpose of evaluation, the method was compared with state of the art methods using the Relative Foreground Area Error (RAE) measurement and the ANOVA analysis of variance tool. The results showed that the proposed method gave the smallest RAE with differ significantly with respect to RAE value compared with other methods, which means that it produced better image results and readability.